We will develop optimal techniques to study clonogenic cells from human brain tumors. The potential problems of clonogenic cell analysis from human tumors include: disaggregation of a rpresentative tumor cell suspension, the influence of the dissociation procedure on subsequent cell survival, tumor heterogeneity, differences between the in vivo and in vitro cell environment, and the low colony forming efficiency (CFE) of cells dissociated from the biopsy. Furthermore, in order to properly interpret human tumor clonogenic cell studies, information must be available on the kinetics of cell proliferation in vitro, the influence of drug treatment duration, and the drug concentration achievable at the tumor cell level in situ. We will consider all of these factors in this proposal. In addition, since tumor cell survival can be influenced by culture conditios, we will optimize our studies for each chemotherapeutic agent to be evaluated. We will determine the in vitro sensitivity of clonogenic cells from malignant gliomas, medulloblastomas, ependymomas, "low grade" astrocytomas, and "normal" brain for agents of defined or theoretical utility for the treatment of patients. Drugs that are administered systemically and intrathecally will be studied. We will compare chemosensitivity between cells derived from tumors of varying degrees of malignancy and with cells derived from "normal" brain. We will correlate cell survival with patient age and initiate investigations on the biochemical differences between tumor cells from young and old patients. Finally, possible interactions of BCNU with polyamine synthesis inhibitors will be investigated.